4.1 Mitochondria
Mitochondria are located in the cytoplasm and play central roles in the maintenance and function of all tissue types. The mitochondrion has a double membrane surrounding a distinct molecular machinery including mtDNA and proteins. The inner mitochondrial membrane (IMM) forms cristae emanating into the matrix and provide a surface for many enzymatic reactions (e.g. lipid metabolism, citric acid cycle, oxidative phosphorylation). Oxidation of substrates generates electrons which pass through enzyme Complexes I to V and leads to ATP synthesis (Figure 8).
The mitochondrial control of energy metabolism is closely linked to cell survival and death.
Permeabilization of the IMM and outer mitochondrial membrane (OMM) precedes the signs of caspase activation leading to apoptosis or necrosis (20,21,271-272). The permeability transition pore (PTP) complex between the IMM and OMM is formed by several proteins including pro- and anti-apoptotic homologs of Bcl-2. The opening and closing of PTP is controlled by the actual MTMP, concentration of divalent cations, matrix pH, oxidation / reduction of nucleotides, and the expression of pro- and anti-apoptotic members of the Bcl-2 family (Figure 19) (20,21,271-272). A complete opening of PTP leads to uncoupling of the respiratory chain, arrest of ATP synthesis, drop of the MTMP, Ca2+ outflow from the matrix, generation of free radicals and a release of pro-apoptotic molecules (cytochrome c, Smac /
dc_5_10
124 DIABLO and apoptosis inducing factor) from the inter-membranous space and matrix to the cytoplasm (20,21,165,221,271-272).
Cytochrome c in the cytosol binds to the apoptotic protease-activating factor 1 (Apaf-1) that leads to the activation of the caspase cascade of apoptosis in the presence of ATP. Without sufficient ATP, the apoptosome cannot be formed and the cell dies by necrosis. Both inherited (e.g. primary LHON mutations) and acquired OXPHOS defects (e.g. oxidative damage) are pore openers leading to the above outlined self-amplifying process and cell death (Figure 19). Activated calcium-dependent
cysteine proteases and caspases also can impair cytoskeletal structures, causing suspension of axonal transport and axonal degeneration (Figure 20).
dc_5_10
125 Figure 19. The role of mitochondria in apoptosis and necrosis
•PTP openers:
• Free radicals
• Oxidative damage
• OXPHOS defects
• Alkalic pH
• Low transmembrane potential
• Bax, Bak
• High Ca++in matrix
•PTP closers:
• Free radical scavengers
• Neutral and acidic pH
• Benzodiazepine agonist
• High transmembrane potential
• Bcl2, BclxL
• Low Ca++in matrix, high external Ca++
Release of cytochrome C into the cytoplasm Activation of Apaf 1 and Caspase 9
Caspase cascade APOPTOSIS
in the presence of ATP in the absence of ATP
Mitochondrial swelling
NECROSIS
In the presence of ATP, the opening of PTP leads to the release of cytochrome c into the cytoplasm, activation of the caspase cascade and apoptosis. In the absence of ATP, PTP opening is followed by mitochondrial swelling and necrosis. PTP openers and closers are listed in the gray windows. Pro-apoptotic members of the Bcl-2 family (Bax, Bak) cause PTP opening, while anti-apoptotic members (Bcl-2, Bcl-xL) are involved in PTP closing.
dc_5_10
126 Figure 20. A proposed mechanism of inflammation induced neurodegeneration in MS
Axonal degeneration in inflammatory demyelination has been related to lack of trophic support (glial cells, growth factors), the loss of protective myelin sheaths, a direct effect of immunoglobulins and cytokines, and excitotoxicity mediated by glutamate and Ca2+. We propose that mitochondria are also involved in the down-stream pathway of neurodegeneration initiated by inflammatory ROS and NO. Oxidative damage to mitochondrial macromolecules may cause OXPHOS impairment, opening of PTP, activation of caspases and apoptosis or other forms of cell death. The activation of caspases and other proteases can also cause a destruction of cytoskeletal proteins.
dc_5_10
127 4.2 A proposed role of mitochondria in inflammation induced neurodegeneration
Based on our observations we propose that inflammation ignites a mitochondrion-driven mechanism that contributes to the observed apoptotic and non-apoptotic oligodendroglial loss and neuroaxonal degeneration in MS (Figure 19,20). Immune mediated forms of oligodendroglial and neuronal apoptosis have been comprehensively studied in MS (55,56,58,59,207,273). A MHC dependent cytotoxicity mediated by CD8 or CD4 T lymphocytes has been debated, since in vivo data do not show significant expression of MHC Class I molecules by oligodendrocytes and neurons, and Class II is not expressed by these cells at all. Nevertheless, CD8 T cells curiously line up along demyelinated axons (63).
However, a by-stander mechanism caused by the engagement of antigen specific T cell receptors with the appropriate MHC molecules on antigen-presenting cells can cause oligodendrocytopathy or
neuronal damage (63). CD95-ligand (L) positive T cells also induce apoptosis of CD95 positive
oligodendrocytes in a non-MHC-dependent manner (59), while activated CD95 positive T lymphocytes are eliminated by CD95L positive residential cells from evolving plaques (274-275). Further, activated macrophages and microglia cause cell damage in the CNS by multiple mechanisms including antibody- and complement-dependent cytotoxicity, cytokine ligand - receptor or adhesion molecule - receptor mediated cytotoxicity (204).
A mitochondrion-driven component of neurodegeneration in inflammatory demyelination had not been described prior to our studies. We propose that this pathway is initiated by inflammation in MS.
Activated monocytes and microglia in MS express iNOS and produce increased amounts of NO (203,279), which damages proteins by generating nitration adducts (e.g. nitrotyrosine). NO can also react with O2- (a component of ROS) resulting in a toxic intermediate called peroxinitrite. ROS are produced in increased amounts by activated MNCs in MS (149) and released by macrophages giving a ring like appearance on MRI in acute plaques (211). ROS cause oxidative damage to macromolecules in the site of inflammation. While evidence suggest that scavengers of NO and physiologic anti-oxidants can provide significant therapeutical effects in EAE (276-278), we did not detect an upregulation of such molecules in lesions of MS. In contrast, we consistently detected a significant accumulation of oxidative damage to DNA in association with inflammation in plaques, by using High Pressure Liquid Chromatography (148), immunohistochemistry and endonucleases (FPG and Endo III) which introduce single strand breaks at oxidized purine and pyrimidine nucleotides, respectively (105). Assaying
mitochondrial enzyme complexes, we found a decreased activity of NADH-DH component of Complex I in 70% of active plaques (105). Despite the accumulated oxidative damage in mtDNA, this decreased Complex I activity was not related to an accelerated accumulation of somatic mtDNA deletions.
Therefore, a direct effect of oxidative stress on the enzyme complex per se may need to be further evaluated. In correlation with markers of immune activation, inflammation, oxidative damage and
OXPHOS impairment, we detected an upregulation of mRNAs for molecules regulating cell survival and
dc_5_10
128 death (136). The temporal and spatial correlations suggest that these events align in a biological sequence and contribute to a mitochondrial regulation of cell survival and death in lesions of MS. While our data had supported alone this concept for years, recent histological analyses with new molecular markers revealed further evidence for an oxidative stress related mitochondrial mechanism of
neurodegeneration in MS (279,280). In these studies, an increased expression and nuclear translocation of hypoxia inducible factor (HIF)-1α was detected in plaques characterized by distal oligodendrocytopathy, apoptotic death of oligodendrocytes and signs of hypoxic tissue injury
(19,279,280). The authors proposed that this condition resulted from oxidative stress and mitochondrial dysfunction, best defined in large acute, histological type III plaques and Balo’s type of concentric lesions, but also present in some chronic active lesions. Subsequent observations on altered mRNA and protein expressions of mitochondrial molecules in MS cortex and white matter lesions can be reconciled with our conclusion and underscore the role of mitochondria in inflammatory demyelination (262-265). However, in addition to inflammation leading to neurodegeneration via a mitochondrial pathway, demyelination may also contribute to a mitochondrial mechanism faciliating axonal loss. This mechanism results from the increased expression and redistribution of the voltage-gaited Na-channels from the node of Ranvier to the entire length of demyelinated axons. The higher numbers of
Na-channels are associated with higher energy demand, but at a time when the Na+/K+ ATPase molecules are compromised by inflammation. This deficiency of the Na+/K+ ATPase leads to an excessive
axoplasmic Ca2+ accumulation via the Na+/Ca++ exchanger. Thus, the chronically depolarized demyelinated axons ultimately suffer from energy depletion, altered Ca2+ homeostasis, and impaired structural integrity (263-265). In addition, both the energy depleted axons and oligodendrocytes are highly sensitive to the toxic effects of glutamate mediated by distinct glutamate receptors (265). The recent identification of a KIF1B SNP as susceptibility marker for MS (116), brings the kinesin family of transport proteins also in the picture. These molecules are responsible for the axoplasmic transport of mitochondria and neurotransmitters, and thus, may also contribute to an inflammation induced, mitochondrion-mediated neurodegenerative process. These data underscore the complexity of
mitochondrial involvement in the final pathway of neurodegeneration in MS. Our works represents the first introduction of a concept that links inflammation via a mitochondrial mechanism to CNS tissue loss.
4.3 The involvement of Complex I in MS
A possible involvement of Complex I in inflammatory demyelination and neurodegeneration was suggested by previous clinical and genetic observations, including 1) the occurrence of inflammatory demyelination in pedigrees with LHON; 2) the detection of primary LHON mutations in mtDNA encoded subunits of Complex I in some patients with MS; 2) the increased frequency of secondary LHON
mutations in mtDNA encoded subunits of Complex I in patients with MS; and 3) the co-localization of several nDNA encoded subunits of Complex I with linkage defined susceptibility loci of MS. These
dc_5_10
129 observations prompted us to perform the most comprehensive analyses of mtDNA to date, and to define variants and haplotypes of nDNA encoded genes of Complex I in MS.
The mtDNA studies were comprehensive in two ways: 1) Screening 25% of the entire mtDNA sequence in populations of patients and controls, and 2) Determining the full sequence of mtDNA in a few patients with MS, PON and NMO. Our studies establish that primary LHON mutations rarely occur, and the involvement of other pathogenic mtDNA mutations also can be excluded in MS, PON and NMO.
However, there is an increased frequency of secondary LHON mutations that align in two MS associated (designated as K* and J*) mtDNA haplotypes in Caucasians. Since primary LHON
mutations predominantly occur in haplotype J*, the observed association between LHON and MS may be related to the overlapping mitochondrial genetic background of the two diseases.
Although the functional significance of secondary LHON mutations remains to be determined, it is notable that these missense variants are all located within mtDNA encoded subunits of Complex I.
Since mtDNA and nDNA encoded subunits of Complex I are assembled in a single protein complex, the small effects of variants in different subunits may become additive or synergistic and influence the overall function of the enzyme complex.
We observed that haplotypes within the NDUFS5, NDUFS7 and NDUFA7 nDNA encoded subunits of Complex I are associated with MS in families. Fathers as well as mothers who carry the MS associated NDUFS5 haplotypes have an increased frequency of the J* mtDNA haplotype including secondary LHON mtDNA variants. This observation suggests an epistatic interaction among nDNA and mtDNA encoded subunits. However, our unpublished observations suggest that the sequence variants don’t directly influence the activity of the enzyme under normal conditions (data not shown). We postulate that the genetic variants only exert very subtle effects on the enzyme’s physical properties, however, these small effects may become additive to or synergistic with the effects of biochemical modifiers (e.g.
inflammatory molecules, oxidative stress) in a pathological environment.
4.4 Summary of conclusions
The limited effectiveness of currently available medications in progressive forms of MS has
reemphasized the importance of neurodegeneration developing in association with inflammation. Loss of oligodendrocytes and failure of their regeneration from precursors is the cause of an irreversible demyelination. A functionally most significant correlate of disability is the neuroaxonal loss, which progresses from the onset of the disease. Direct inflammatory injury contributes to focal loss of axons, neurons and oligodendrocytes within circumscribed lesions, while subtle degenerative processes are likely more instrumental in the diffuse loss of these tissue elements outside of plaques. We propose a
dc_5_10
130 concept of neurodegeneration in MS as a process linked to inflammation via an acquired mitochondrial dysfunction and to its consequences including apoptotic and non-apoptotic cell death. Individual differences in the tissue response to inflammation show great variations and may depend on genetic polymorphisms in molecules defining the mitochondrial regulation of energy metabolism, cell survival and death. Earlier MS genetic studies identified several susceptibility loci in chromosomal regions where subunits of OXPHOS enzymes are encoded. Our early studies revealed an association between mtDNA variants in subunits of Complex I, cytochrome b and MS. The subsequent studies on nuclear genes of Complex I demonstrate that nucleotide variants and haplotypes within the genes of NDUFS5, NDUFS7 and NDUFA7 subunits are associated with the disease in families, and that an interaction between mtDNA and nDNA variants of Complex I may occur at protein level. Our biochemical and histological data suggest that Complex I may be affected by oxidative stress and contribute to the degenerative process downstream to chronic inflammation in the CNS. While targeting inflammatory mediators (e.g. CCL-CCR molecules) shortly after the onset or early during the course of the disease is necessary to prevent the development of a downstream pathology (demyelination,
oligodendrocytopathy, neuroaxonal loss), new strategies (e.g. neuroprotection preserving mitochondrial function) are needed to minimize neurodegeneration in both early and established forms of the disease.
Identification of genetic markers rendering individuals susceptible to autoimmunity and
neurodegeneration associated with inflammation may facilitate the development of an early combined therapy personnaly designed for the treatment of multiple sclerosis.
dc_5_10
131
Acknowledgments:
I am greatly indebted to Drs. Andras Guseo, Gyozo Petranyi and Peter Halasz for the exceptional professional and personal support they provided early during my studies in neurology, MS and immunogenetics, which allowed me to develop interest in research. I am also very grateful to Dr. Samuel Komoly for his collaboration and
continuing support in building bridges with the Medical University of Pecs.
I am very grateful to Dr. George C. Ebers and Dr. Jorge Oksenberg for allowing us obtaining DNA specimens for the family based genetic studies from the Collection of the Canadian MS Collaborative Group, London, ON, and from the Multiple Sclerosis DNA Bank, UCSF, San Francisco, CA, respectively. Dr. Gary Birnbaum also
contributed specimens of PP-MS families from the Multiple Sclerosis Treatment and Research Center,
Minneapolis, MN. Specimens from sporadic patients with MS and controls were collected with the help of Dr. Fred D. Lublin, at the Multiple Sclerosis Center, Thomas Jefferson University, Philadelphia, PA.
Peers who contributed to my studies (in chronological order):
Michael D. Brown, Ph.D. discussed mitochondrial genetics of LHON.
Hansjuerg Alder, Ph.D. consulted technical issues in the mtDNA studies.
Fred D. Lublin, M.D. invited me for a one year fellowship in the USA, and contributed specimens to my studies.
Raul N. Mandler, M.D. contributed specimens to the studies on Devic’s disease.
George C. Ebers, M.D. generously allowed me to learn about linkage analysis during my sabbatical at the Wellcome Trust Center, Oxford University.
Yin Yao Shugart, Ph.D., a genetic statistician, consulted some aspects of data analyses.
Mary Selak, Ph.D. was involved in the assessment of Complex I activity.
Dr. Saud Sadiq generously offered space for moving my lab from Philadelphia to the Roosevelt Hospital, Columbia University, New York, NY and helped us with his patients’ specimens.
Postdoctoral fellows, associates and technicians trained in my lab and involved in my projects:
Shulan Li, Ph.D, Mary R. Vohl, Ursula Bosch, M.D, Jose L. Rodriguez, M.D., Alan Cahill, Ph.D., Fengmin Lu, M.D., Ph.D., Linda Shawver, John O’Connor, Ph.D., Olga Vladimirova, Ph.D., Devjani Chatterjee, Ph.D. and Zheng Kee, M.D., Ph.D. were involved in the laboratory procedures or in running the computer programs used in the mitochondrial DNA and biochemical studies. Tamara Vyshkina, Ph.D. ran the computer programs in the family based association studies. Ileana Banisor carried out the laboratory procedures in the real-time PCR studies.
Andrei Blokhin and was involved in the somatic mtDNA deletion studies as a postdoc in my lab.
At last, but not least, I will never be able to say enough thanks to my parents who unconditionally and with so much love supported my personal and academic progress from the earliest age to date.
dc_5_10
132
The studies presented here, including all salaries, laboratory equipments, contract and consultation fees, laboratory supplies and overhead fees, were supported by competitive external grants to the author (as principal investigator) of this thesis from the following agencies:
1. Characterization of the mitochondrial DNA in multiple sclerosis. RG2770A2/2 National Multiple Sclerosis Society. $ 321,379. 1996-2000.
2. Analysis of subcellular pathology in the brain of MS patients. PP0501 National Multiple Sclerosis Society. $24,681 1996-1997.
3. Mitochondrial DNA variants in Devic's disease.
National Multiple Sclerosis Society, PP0765 $27,500. 2000-2002.
4. Traveling Grant. The Burroughs Wellcome Fund. $12,344. 2000-2001.
5. Nuclear and mitochondrial candidate genes in multiple sclerosis. RG3334-A-4/T National Multiple Sclerosis Society, $405,594. 2002-2005.
6. Candidate genes in primary progressive multiple sclerosis.
Wadsworth Foundation $330,000. 2002-2005.
7. The role of genetic variations in β-chemokines in defining the disease modifying effect of Rebif in MS. Serono
$120,000. 2003-2005.
8. Variants of beta-chemokines within chromosome 17q11 in MS. RG35212-A-6 National Multiple Sclerosis Society $632,251. 2004-2006.
9. Mitochondrial DNA deletions in MS brains. NMSS. PP1334 $44,000, 11.01.2006-10.31.2008.
dc_5_10
133
CMSCG: Canadian Multiple Sclerosis Collaborative Group CNS: central nervous system
HBSFRC: Human Brain and Spinal Fluid Resource Center HIF1α: Hypoxia inducible factor-1α
HIPAA: Health Insurance Portability and Accountability Act HLA: human leukocyte antigen
HPLC: high pressure liquid chromatography HS: heavy strand (of mtDNA)
HWE: Hardy-Weinberg equilibrium
IDDM: insulin dependent diabetes mellitus IMM: inner mitochondrial membrane
MTR / MTI: magnetization transfer ratio / imaging NAA: N-acethyl-aspartate
NAGM: normal appearing gray matter NAWM: normal appearing white matter
dc_5_10
134
NCBI: National Center for Biotechnology Information nDNA: nuclear DNA
NMO: neuromyelitis optica NO: nitric oxide
OMM: outer mitochondrial membrane ON: optic neuritis
OND: other neurological disease OXPHOS: oxidative phosphorylation PBL: peripheral blood lymphocyte PCR: polymerase chain reaction PDT: pedigree disequilibrium test PON: prominent optic neuritis PP-MS: primary progressive MS PR-MS: progressive-relapsing MS PTP: permeability transition pore QTL: quantitative trait locus RA: rheumatoid arthritis
RMMSBB: Rocky Mountain Multiple Sclerosis Brain Bank rRNA: ribosomal RNA
ROS: reactive oxygen species RR-MS: relapsing-remitting MS RP-MS: relapsing-progressive MS SLE: systemic lupus erythematosus SNP: single nucleotide polymorphism SOD-1: superoxide dismutase-1 SP-MS: secondary progressive MS TCR: T cell receptor
TDT: transmission disequilibrium test TH1 / TH2: T helper-1 / T helper-2 tRNA: transfer RNA
UCSF MSDB: University of California, San Francisco, Multiple Sclerosis DNA Bank VEP: visual evoked potential
VCAM1: vascular cell adhesion molecule-1 VLA4: very late antigen-4
WM: white matter
dc_5_10
135
References:
1. Kalman B, Lublin FD. The genetics of multiple sclerosis. A review. Biomed Pharmacother 1999;53:358-370.
2. Kurtzke JF, Beebe GW, Norman JE, Jr. Epidemiology of multiple sclerosis in U.S. veterans: 1. Race, sex, and geographic distribution. Neurology 1979;29:1228-1235.
3. Pratt RTC, Compston ND, McAlpine D. The familial incidence of multiple sclerosis and its significance.
Brain 1951;74:191-232.
4. Bulman DE, Ebers GC. The geography of multiple sclerosis reflects genetic susceptibility. J Trop and Geograph Neurol 1992:66-72.
5. Ebers GC, Bulman DE, Sadovnick AD, Paty DW, Warren S, Hader W, Murray TJ, Seland TP, Duquette P, Grey T, et al. A population-based study of multiple sclerosis in twins. New Engl J Med 1986;315:1638-1642.
6. Ebers GC, Sadovnick AD, Risch NJ. A genetic basis for familial aggregation in multiple sclerosis.
Canadian Collaborative Study Group. Nature 1995;377:150-151.
7. Ebers GC, Yee IM, Sadovnick AD, Duquette P. Conjugal multiple sclerosis: population-based prevalence and recurrence risks in offspring. Canadian Collaborative Study Group. Ann Neurol
7. Ebers GC, Yee IM, Sadovnick AD, Duquette P. Conjugal multiple sclerosis: population-based prevalence and recurrence risks in offspring. Canadian Collaborative Study Group. Ann Neurol